Data considerations for Accelerated Life Testing

To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.

The response variable should be continuous
Continuous data are measurements that may potentially take on any numeric value within a range of values along a continuous scale, including fractional or decimal values. For an accelerated life test, the continuous response is often defined as the failure time for each unit at one or more stress levels.
If you have binary response data with only two possible outcomes, such as fail vs not fail, use Probit Analysis.
You must account for censored data

Life data are often censored, which means that the exact failure times of some items are unknown. If you have censored observations, you must include them in your analysis to obtain accurate reliability estimates.

Use right-censoring to credit success time to items that have not yet failed. Use interval- or left-censoring to account for uncertainty when you don’t know the exact failure times. For more information, go to Data censoring.

You must have at least one accelerating variable
You can have two predictors for an accelerated life test, but at least one predictor must be an accelerating variable. The second predictor can be either a second accelerating variable or a factor.
For example, an engine with an average life of many thousands of hours when run under a standard speed might be tested at an accelerated stress level of twice this speed. The time to failure at the normal-use conditions can be extrapolated from the elevated stress level. Base the stress levels on your knowledge of the system being tested. Stress levels that are too high may break down a system suddenly instead of simply accelerating wear. For example, stress testing the compression strength of a paper egg carton at several thousand pounds of pressure causes a nearly immediate breakdown and offers little insight into the strength of the carton under normal conditions. For more information on stress levels, go to Stress levels for accelerated life test plans.
The model that you use must fit the data
If the fitted model that you select does not adequately represent the data. the results will not be accurate. Use engineering knowledge about the relationship between failure time and the accelerating variable to choose the appropriate model. The probability plots in the results can help you to assess whether the distribution, the relationship for the accelerating variable, and the assumption of equal shape (or scale) are appropriate at levels of the accelerating variable. However, engineering knowledge is the only way to verify that the model is appropriate at design temperatures.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy